We are developing special elictation tasks to gather rich, matched data cross multiple languages. Our aim is to understand the kinds of grammatical solutions that are available to coordinate the attention and belief of a speaker and a hearer, and how shared knowledge is monitored.
As part of this we are developing a semi-parallel corpus of around fifteen different languages from around the world that differ dramatically in their structures. The data collected in this research strand will form a semi-parallel, naturalistic corpus for how languages encode categories relevant to social cognition.
In this research we zero-in on particular communicative functions and compare their structures across a broad range of languages. This research will inform debates on how far the shared, fundamental human capacity for social cognition is modulated by different linguistic structures.